Invasive cervical cancer is the 4th most common cancer and cause of cancer-related mortality in women worldwide. Low- and middle-income countries (LMICs) experience almost 90% of the global cervical-cancer burden, with sub-Saharan Africa (SSA) experiencing the highest rates of cervical cancer. Human immunodeficiency virus (HIV) infection, which also disproportionately affects SSA, greatly increases the risk of cervical cancer. The World Health Organization now has a call-to-action for the elimination of cervical cancer, which includes vaccination young adolescents against human papillomavirus (HPV), the obligate viral cause of cervical cancer, and screening mid-adult women for the early detection and treatment of cervical abnormalities before becoming cancerous. However, both the best strategy for treating cervical abnormalities in women living with HIV (WLWH) from SSA and how to mitigate treatment failures are unknown. To address this gap in knowledge, we propose a randomized clinical trial to assess the treatment effectiveness of two ablative methods of treatment, gas-based cryotherapy and thermocoagulation, for treatment of CIN2/3 and high-risk HPV in WLWH women. Eligible, confirmed WLWH women (n=5,014), ages 25-49 years, and living in Maputo, Mozambique, in conjunction with the local PEPFAR (President's Emergency Plan For AIDS Relief) program, will be recruited to participate in this trial when attending their routine HIV-care visit. Consenting WLWH will be screened by rapid, point-of-care hrHPV DNA testing, unaided visual inspection after acetic acid (VIA), and ?deep learning?-derived, automated visual evaluation (AVE) algorithm applied to a digital image captured on a cell phone, thereby ensuring that most CIN2/3 in the population is identified and treated. Screen-positive (HPV, VIA, 6-mo AVE positive) women will be 1) randomly assigned to either the GBC or thermocoagulation, 2) undergo a rigorous colposcopic evaluation and biopsies to determine the (post-hoc) diagnosis of the cervical abnormality, and 3) undergo their assigned treatment if ablation eligibility or LEEP if ineligible based on WHO guidelines8. Adverse events and pain data will be collected systematically during the treatment visit and during an at-home, one month post-treatment visit by community health workers. Six- and 18-month follow-up visits will be used to assess effectiveness of these treatments against CIN2/3 and hrHPV. Nested studies will evaluate whether 1) specific HPV genotypes, CD4 and HIV viral load, and/or lesion size, location, and severity (i.e. imperfect adherence to WHO guidelines), are risk factors for treatment failures by each method and 2) deep-learning algorithms applied to digital images can predict ablative treatment failures and thereby used in the future to triage screen-positive women to LEEP instead of ablation and thereby potentially be avoided those treatment failures. Secondary analyses will assess the effectiveness of 12 different S&T strategies based on different combinations of 6 screening methods (hrHPV with no triage, hrHPV with triage using HPV genotype groups, hrHPV with VIA triage, hrHPV with AVE triage, VIA alone, or AVE alone) and two ablative methods.